Assessment of Factors that Contribute to Decreased Quality of Life in Gynecologic Oncology Group Ovarian Cancer Trials

Abstract

Objective

To assess which QOL line items are associated with low QOL in women receiving chemotherapy for ovarian cancer.

Methods

Patients with stage III or IV ovarian cancer on Gynecologic Oncology Group protocols 152 and 172 who underwent primary surgery followed by intravenous paclitaxel and cisplatin completed the Functional Assessment of Cancer-Therapy Ovarian (FACT-O). The FACT scale includes the 4 domains of physical, functional, social, and emotional well-being (PWB, FWB, SWB, EWB). Women whose overall FACT-O score was in the lowest quartile (Q1) were compared with women in the upper three quartiles (Q2-4). The proportions of women in each group who selected the two worst categories for each item were compared. Significance was set at p<.005.

Results

Prior to cycle 4, 361 (86.4%) patients provided valid QOL assessments. For PWB, a significantly higher proportion of women in Q1 versus Q2-4 selected the two worst categories of several physical symptoms (nausea, pain, feeling ill and being bothered by the side effects of treatment). For FWB, significant differences included being able to work, being content with the quality of their life, and sleeping well. For EWB there were significant differences in feeling nervous and worrying about dying. There were virtually no differences between groups in SWB. Low interest in sex was reported by 56-88% of all patients (Q1-Q4).

Conclusions

A large proportion of women whose total FACT-O score is in the lowest quartile are reporting problems that may be amenable to clinical interventions such as symptom management and psychosocial support.

Introduction

Quality of life (QOL) assessments are a key factor in assessing cancer burden and the effect of treatment. In a single institution study, QOL was measured in ovarian cancer patients during adjuvant chemotherapy [1]. Mean scores of physical well-being and fatigue were significantly lower post-operatively and during adjuvant therapy but improved to pre-operative levels following chemotherapy. Mean functional well-being subscale scores, however, increased above pre-operative levels following chemotherapy. Emotional and social well-being subscale scores did not change. Wenzel et al found significant decreases in total QOL scores measured during chemotherapy, primarily due to changes in physical and functional well-being [2]. From these studies [1,2] it appears that overall QOL is decreased in ovarian cancer patients during chemotherapy, especially in the physical domain.

Average QOL assessment scores do not provide information on which specific items within a domain may be compromised. Within the physical well-being domain, fatigue, nausea, pain, side effects of treatment, malaise, performance status and ability to meet family needs are the individual items. The degree to which individual items contribute to the overall Functional Assessment of Cancer Therapy (FACT) score has not been studied in the cancer population. This has practical and clinical importance, as some of these items indicate symptoms or problems which may be treated. If impairment in total QOL is primarily due to decrements in specific items, treatment of these items are likely to improve QOL [3-8]. Reductions in distress, and improvement in overall QOL may also influence treatment outcomes, including survival [2]. Wenzel et al [2] found that baseline QOL was predictive of survival in ovarian cancer patients, and this observation was primarily attributed to patients in the lowest scoring quartile.

The primary objective of this study was to assess which items are associated with lower total FACT-Ovarian (FACT-O) scores obtained during chemotherapy by comparing domain item responses from patients in the lowest quartile with those in the upper three quartiles. Previous work has demonstrated that women whose overall QOL score is in the lowest quartile may differ significantly from women in the upper quartiles [2] and examining women in the lowest quartile may also increase the clinical relevance of the data. As inclusion of QOL scores are becoming accepted practice in outcome assessment [9] appreciation of the range of scores typical for a given disease will increase, as will realization of when women are likely to be in the lowest quartile of normative scores for that condition. The primary hypothesis was that women whose overall QOL was in the lowest quartile would select specific items as being significantly more problematic for them than would women whose overall QOL was in the upper three quartiles. The null hypothesis was that women in the lowest quartile would have a generalized depression of scores and that they would not tend to identify specific items that were significantly worse for them.

QOL data during chemotherapy are available from two national Gynecologic Oncology Group (GOG) studies (GOG 152 and 172), which both included intravenous (IV) chemotherapy study arms. GOG 152 enrolled suboptimally debulked women while GOG 172 enrolled optimally debulked women. Whereas long term prognosis for the two groups may differ, they had both received the same treatment at the time of this assessment. Including both of these protocols provided an opportunity to examine whether debulking status influenced the domain items with decreased scores, and increased the generalizability of the data. Specifically, it was expected that debulking status would not affect the majority of the items within the QOL domain, but rather, the primary hypothesis would be observed in both groups.

Methods

Patients

Data for this analysis were obtained from patients enrolled on the control arms of two Gynecologic Oncology Group (GOG) randomized ovarian cancer protocols. All patients enrolled in GOG 152 [2] and GOG 172 [9] who were randomized into the 6 cycle intravenous (IV) paclitaxel and cisplatin control arms were included (identical dosing). GOG 152 measured QOL in a randomized trial of interval secondary cytoreduction in suboptimally debulked women with advanced ovarian carcinoma. GOG 172 measured QOL in a randomized study of IV paclitaxel and cisplatin versus IV paclitaxel, IP cisplatin and IP paclitaxel in optimally debulked stage III epithelial ovarian cancer patients. GOG 152 enrolled 208 patients with no interval cytoreduction and GOG 172 included 210 patients entered into the IV cisplatin and paclitaxel arm. Patients receiving IP therapy on GOG 172 and patients who had secondary debulking surgery on GOG 152 were not included in the analysis. Participating institutions obtained institutional review board approval of the protocols before enrolling any patients; all patients provided written informed consent consistent with all federal, state, and local requirements before they received any protocol therapy. Treatment information and outcome has been previously reported [10,11].

Methods

For the current report, the baseline QOL measure in GOG 152 (prior to cycle 4 of chemotherapy) and the second QOL measurement from GOG 172 (prior to chemotherapy cycle number 4) were analyzed. The QOL measure chosen (prior to chemotherapy cycle 4, after the third cycle) was used in order to provide an assessment of QOL during mid-treatment chemotherapy after several cycles have already been administered.

QOL was measured with the FACT-O questionnaire. The FACT-O scale includes the FACT-General (FACT-G) along with 12 additional questions specific to ovarian cancer patients. The FACT-G, version 4, is a 27-item core questionnaire evaluating the domains of physical, functional, family-social, and emotional well-being (PWB, FWB, SWB, EWB) [12]. Questions are answered on a 5-point Likert scale (0 = not at all; 1 = a little bit; 2 = somewhat; 3 = quite a bit; and 4 = very much) and a subscale (with N_i items) score S_i was computed as follow when 50% or more of the items had been answered:

$$S_i=N_i\times \frac{\sum ({\delta }_{ij}\times s_{ij})}{\sum ({\delta }_{ij})}$$

where δ_ij is equal to 1 when the j^th item has a valid response; otherwise it is equal to 0 and s_ij is the score of the j^th item. The FACT-O score is the sum of the five subscale scores (FWB, EWB, SWB, PWB and Ovarian specific subscale) and is considered valid only when more than 80% of the items are complete. The range of the FACT-O score is 0 – 156 and higher FACT-O scores are associated with better QOL. Reliability, validity and sensitivity to change of the FACT-G have been demonstrated in a variety of settings and relative scores can be compared to normative data [13]. Basen-Engquist et al demonstrated reliability and validity of the FACT-O instrument in ovarian cancer patients with internal consistency reliability for the subscales ranging from 0.74 to 0.92 and test-retest reliability 0.72 to 0.88 [14]. The EWB score in GOG 152 only included 5 items and this domain score was prorated by assuming the sixth item was missing [15].

Total FACT-O scores for all patients were calculated. Patients were divided into quartiles according to their total FACT-O score (Q1-Q4). Patients in the lowest quartile (Q1) were compared to women in the upper three quartiles (Q2-4). This comparison was chosen to expand upon Wenzel et al's observation that baseline QOL was predictive for survival and was primarily attributed to the lowest-scoring quartile [2]. Patient demographic and clinical characteristics (age, race, ethnicity, stage and performance status) between patients in Q1 versus Q2-Q4 were compared by use of Student's t-test or chi-square test of proportions. Performance status is a scale used to assess how a patient's disease affects the daily living ability of the patient (0=fully active, 1=restricted in physically strenuous activity, 2=ambulatory and capable of all self-care but unable to carry out work activities, 3=capable of limited self-care) [16]. Performance status was available at different time points for the two studies; prior to cycle 1 (at baseline) for patients on GOG 172 and prior to cycle 4 for GOG 152.

Each of the 39 items that made up the FACT-O scale were examined for patients in Q1 versus Q2-4. Comparisons were done separately for GOG 152 and 172 studies as the two protocols differed in terms of debulking status. The proportion of women who selected the two worst response categories for each item was compared between the two groups (Q1 versus Q2-4). It is important to note that item response sets vary (e.g., “I am able to work” or “I have a lack of energy.”) Thus, the two worst categories represent “not at all” and “a little bit” for the positive items, and represent “very much” and “quite a bit” for the negative items. Comparisons were made by use of the chi-square test of proportions and only p values < 0.005 were considered statistically significant due to the number of comparisons made.

Results

There were 208 eligible suboptimally debulked patients enrolled on the arm of 6 cycles of cisplatin and paclitaxel without interval debulking in GOG 152 from June 6, 1994 to January 29, 2001. There were 210 eligible optimally debulked patients enrolled for 6 cycles of IV cisplatin and paclitaxel arm in GOG 172 from March 23, 1998 to January 29, 2001. Prior to cycle 4, 189 patients (91%) in GOG 152 and 172 patients (82%) in GOG 172 provided valid QOL assessments for all subscales of FACT-O. Demographic and clinical characteristics of patients on GOG 152 and 172 were similar (Table 1). Mean age of patients was 57 years old, the majority were non-Hispanic Caucasian, almost all had stage III disease and a performance status of 0 or 1. Women were categorized according to either FACT-O ≤ Q1 or FACT-O > Q1. Patients with FACT-O scores in Q1 were not significantly different from those in Q2-Q4 for both protocols with regard to patient characteristics (data not shown).

Table 1. Patient Demographics from GOG 152 and 172.

	GOG #152 (n=189)		GOG #172 (n=172)		P*
Characteristic	No	%	No	%
Age (years)
<40	12	6.3	11	6.5
40-49	34	18.0	32	18.9
50-59	62	32.8	59	34.9
60-69	57	30.2	47	27.8
>=70	24	12.7	20	11.8
Mean (SD)		57.1 (10.9)		56.7 (11.0)	.729
Race
White	167	88.4	150	88.8	.920
Other	22	11.6	19	11.2
Ethnicity
Non-Hispanic	185	97.9	160	94.7	.105
Hispanic or Latino	4	2.1	9	5.3
Stage
III	183	96.8	169	100	.053
IV	6	3.2	.	.
Performance Status
0	107	56.6	73	43.2	.021
1	73	38.6	90	53.3
2	9	4.8	6	3.6

GOG, Gynecologic Oncology Group; SD, standard deviation

^*Student's t-test or chi-square test for comparison of proportions

When dividing patients in to quartiles based on QOL responses, patients in the lowest quartile (Q1) of the FACT-O had total QOL scores that were ≤106.1 in those with suboptimally debulked disease (GOG 152) and ≤102.4 in those with optimally debulked disease (GOG 172) (Table 2). In PWB, a significantly higher proportion of women selected the two worst categories of several physical symptoms (nausea, pain, feeling ill and being bothered by the side effects of treatment) as well as more general effects (lack of energy, meeting needs of family, forced to spend time in bed, Table 2). In terms of functional well being, a significantly higher proportion of women in Q1 selected the two worst categories of being able to work, finding work fulfilling, being able to enjoy things and being content with the quality of their life. A significantly higher proportion selected the worst categories of sleeping well. In regards to EWB, a significantly higher proportion of women in Q1 reported feeling nervous, sad and worrying about dying versus women in Q2-4 on both protocols. For SWB significant differences between patients in Q1 versus Q2-4 were feeling close to my friends for patients on GOG 152 and being satisfied with their sex life for both protocols. Over half the patients in Q1 on GOG 152 and 172 chose the lowest categories of being satisfied with their sex life (SWB). Half to three-quarters of all patients (Q1-Q4) in both protocols chose the lowest categories of being interested in sex (FACT-O items). The FACT-O items that differed significantly between women in Q1 versus Q2-4 on both protocols were have a good appetite, like the appearance of my body, and able to feel like a woman.

Table 2. Percentage of patients who selected the worst two item categories (Q1 versus Q2-4).

FACT-O Items		GOG 152 (n=189)			GOG 172 (n=172)
		FACT-O ≤106.1 (N=47)	FACT-O >106.1 (N=142)	Difference between Q1 and Q2-4	FACT-O ≤102.4 (N=43)	FACT-O >102.4 (N=129)	Difference between Q1 and Q2-4
Physical Well-Being	I have a lack of energy	60%	12%	48% **	77%	24%	53% **
	I have nausea	23%	3%	20% **	30%	6%	24% **
	I have trouble meeting the needs of my family	34%	1%	33% **	52%	9%	43% **
	I have pain	13%	2%	11% **	17%	3%	14% *
	I am bothered by side effects of treatment	43%	5%	38% **	56%	12%	44% **
	I feel ill	23%	.7%	22% **	30%	3%	27% **
	I am forced to spend time in bed	23%	1%	22% **	35%	.8%	34% **
Social Well-Being	I feel close to my friends	23%	1%	22% **	12%	2%	10%
	I get emotional support from my family	.0%	1%	1%	.0%	.0%	0%
	I get support from my friends	2%	7%	5%	2%	.8%	1%
	My family has accepted my illness	4%	1%	3%	.0%	.0%	0%
	I am satisfied with family communication about my illness	9%	5%	4%	5%	.8%	4%
	I feel close to my partner	6%	3%	3%	8%	3%	5%
	I am satisfied with my sex life	59%	23%	36% **	52%	22%	30% **
Emotional Well-Being	I feel sad	28%	2%	26% **	21%	4%	17% *
	I am satisfied with how I am coping with my illness	13%	2%	11% **	21%	6%	15%
	I am losing hope in the fight against my illness	2%	4%	2%	5%	.8%	4%
	I feel nervous	45%	8%	37% **	23%	2%	21% **
	I worry about dying	30%	4%	26% **	19%	2%	17% **
	I worry that my condition will get worse	.	.		28%	2%	26% **
Functional Well-Being	I am able to work	49%	18%	31% **	63%	23%	40% **
	My work is fulfilling	51%	13%	38% **	68%	14%	54% **
	I am able to enjoy life	28%	4%	24% **	40%	5%	35% **
	I have accepted my illness	9%	3%	6%	19%	5%	14%
	I am sleeping well	36%	8%	28% **	49%	11%	38% **
	I am enjoying the things I usually do for fun	57%	16%	41% **	70%	20%	50% **
	I am content with the quality of my life right now	61%	13%	48% **	63%	15%	48% **
Ovarian Cancer Subscale	I have swelling in my stomach area	15%	2%	13% *	5%	.8%	4%
	I am losing weight	17%	2%	15% **	19%	5%	14%
	I have control of my bowels	17%	9%	8%	26%	8%	18% *
	I have been vomiting	9%	.0%	9%	9%	2%	7%
	I am bothered by hair loss	60%	31%	29% **	38%	25%	13%
	I have a good appetite	32%	8%	24% **	34%	9%	25% **
	I like the appearance of my body	62%	23%	39% **	60%	20%	40% **
	I am able to get around by myself	.0%	.7%	1%	2%	.8%	1%
	I am able to feel like a women	36%	3%	33% **	29%	10%	19% **
	I have cramps in my stomach area	15%	3%	12%	21%	4%	17% *
	I am interested in sex	78%	60%	18% **	64%	58%	6%
	I have concerns about my ability to have children	2%	.8%	1%	.0%	.0%	0%

^*p <0.005 for comparison of Q1 vs. Q2-Q4 for GOG 152 or GOG 172

^**p <0.001 for comparison of Q1 vs. Q2-Q4 for GOG 152 or GOG 172

Q1, quartile 1; Q2-4, quartiles 2-4; GOG, Gynecologic Oncology Group, FACT-O, Functional Assessment of Cancer Therapy-Ovarian

Comparisons between women in Q1 of GOG 152 and Q1 of GOG 172 demonstrated a non-significantly higher proportion of patients in Q1 of GOG 152 (suboptimally debulked) selected the worst categories of feeling nervous than did patients in Q1 from GOG 172 (optimally debulked) (45% versus 23%; p=.017). No significant differences in domains or items between Q1 of GOG 152 and Q1 of GOG 172 were observed.

Discussion

For women whose total QOL was in the lowest quartile (Q1), our results demonstrate that the decreases are due to sizeable differences in specific concerns including pain, nausea, feeling ill and being bothered by the side effects of treatment. These data are significant in that several of these items may be amenable to interventions that may then increase other aspects of physical well-being such as having more energy and being able to meet the needs of family. Future research should also examine whether interventions that ameliorate pain, nausea, feeling ill and being bothered by side effects of treatment improve functional well-being. The fact that there were virtually no differences in social well being between these subgroups of women supports the primary hypothesis that impairments in QOL are specific to the disease and treatment and do not reflect a generalized impairment of QOL. Relatively minor interventions that improve physical symptoms may have large effects by improving other aspects of physical and functional well-being.

In this study anxiety appears to be more problematic than depression in women with ovarian cancer. A higher proportion of women who were suboptimally debulked selected the worst categories of feeling nervous (EWB) relative to women who were optimally debulked (45% versus 23%). Other studies have also found a higher incidence of anxiety over depression in ovarian cancer patients [17]. However, the baseline QOL measure in the suboptimal group was done before randomization to possible further surgery. Therefore, the increase in anxiety could be due to patients knowing their suboptimal tumor status or to waiting for randomization. Nevertheless, it appears that a suboptimal debulking is a burden not only on overall survival but also on emotional well-being of the ovarian cancer patient. While many of the disruptions in QOL appear to be due to effects of chemotherapy, specific characteristics of the disease (such as prognosis), may affect overall QOL during treatment. Again, these data are significant in that they identify potential areas of intervention (such as counseling) to alleviate some of these feelings and concerns, especially among women who are suboptimally debulked.

Fatigue compromises QOL in cancer patients. Medical conditions and treatments sometimes contribute to fatigue, including anemia, radiation, chemotherapy, and medication. In addition, other symptoms, including insomnia, pain, and depression may also compound fatigue [18]. Physicians should screen for fatigue on a regular basis, and search and treat secondary causes. Evidenced-based recommendations for fatigue remain elusive as exercise, energy conservation, counseling, and pharmacotherapy are not strongly supportive [19].

Sexual problems are prevalent among ovarian cancer patients. Taylor et al assessed sexual function in ovarian cancer patients who were receiving treatment, off treatment with disease, and those without evidence of disease [20]. The study revealed that those receiving chemotherapy had the lowest sexual activity. Liavaag et al conducted a cross-sectional study which found a lower level of sexual activity in ovarian cancer patients compared to age-matched controls from the general population [17]. Therefore, both studies support that sexual changes appear to be related to active chemotherapy. The latter study also found that sexually active patients reported significant alterations in pleasure and discomfort [17]. This study demonstrates that regardless of whether sexual activity is being assessed in terms of satisfaction (SWB), or with respect to interest (FACT-O items), the majority of patients are having difficulty in this area.

Research limitations of this study include an ancillary analysis of previously collected data with limited patient demographics. Variables such as education, socioeconomic status were not collected and may be important variables when comparing QOL. In addition, GOG 152 and 172 used cisplatin and paclitaxel IV chemotherapy while carboplatin and paclitaxel are currently the preferred regimens used. Cisplatin has been shown to have more toxicity than carboplatin. It is possible that line items identified with lower QOL would differ in patients who received carboplatin and paclitaxel as compared to those identified in this study.

Women with a low QOL should be evaluated to determine if clinical interventions are appropriate. Further impetus for evaluation of these women is that QOL at baseline is predictive of survival in ovarian cancer patients, suggesting that these impairments may not only be disruptive of QOL but may have serious long term consequences [21]. We hypothesize that by targeting and improving specific QOL areas that there could be global improvements thereby having the potential to improve overall survivorship.